Difficulty: Introductory | Prerequisites: Storage Hardware Fundamentals study notes (Part 1)
Once you understand the physical technologies that store data (hard drives, SSDs, flash, optical), the next question is how those technologies are connected, combined, and scaled. This material covers network and cloud storage, smart cards, RAID configurations, archival systems, and the criteria for choosing between storage options. These topics bridge the gap between a single device on your desk and the large-scale infrastructure that powers data centres, cloud services, and enterprise IT. If your course moves into networking, security, or systems administration, the concepts here are prerequisites.
Storage can live on a local network (NAS, SAN) or in the cloud. RAID combines multiple disks for speed, redundancy, or both. Archival systems prioritise cost and durability over access speed. Choosing the right storage means weighing speed, capacity, compatibility, portability, and cost.
Remote storage
Any storage that is not directly attached to the user's computer. Data is accessed over a network rather than through a local cable or internal connection.
In simple terms, the drive is somewhere else and you reach it through a network.
Network-Attached Storage (NAS)
A dedicated storage system connected to a local area network (LAN), allowing multiple users and devices to access shared files.
Think of it as a shared hard drive that everyone in the office (or house) can read and write to over Wi-Fi or Ethernet.
Storage Area Network (SAN)
A specialised, high-speed network that connects servers to a pool of storage devices. SANs are used in enterprise and data-centre environments where large volumes of data must be accessed quickly and reliably.
Think of it as a private motorway built just for storage traffic, separate from the regular office network.
Cloud storage
Storing data on remote servers managed by a third-party provider and accessed via the internet. Google Drive, Dropbox, and OneDrive are common examples.
In simple terms, your files live on someone else's servers and you get to them through a browser or app.
Smart card
A credit-card-sized plastic card with an embedded microprocessor and memory, capable of storing up to about 64 KB of data. Used for identification, access control, prepaid transactions, and secure authentication.
RAID (Redundant Arrays of Independent Disks)
A technology that combines multiple physical disks into a single logical unit to improve performance, fault tolerance, or both.
Think of it as teamwork for hard drives: several disks sharing the load so that data is faster to access, safer from a single-disk failure, or both.
RAID 0 (disk striping)
Data is split across multiple disks so reads and writes happen in parallel. This improves speed but provides no redundancy: if one disk fails, all data is lost.
RAID 1 (disk mirroring)
Data is duplicated identically on two disks. If one disk fails, the other still has a complete copy. The trade-off is that you lose half your total disk capacity to the mirror.
Fault tolerance
The ability of a system to continue operating correctly even when one or more components fail.
Archival storage
Long-term storage for data that is inactive or rarely accessed. Archival systems prioritise low cost and durability over fast retrieval.
Fibre Channel
A high-speed networking protocol used primarily in SANs to connect servers to storage. Offers very low latency and high throughput.
iSCSI (Internet Small Computer Systems Interface)
A protocol that carries SCSI storage commands over standard TCP/IP networks, allowing storage to be accessed remotely without specialised Fibre Channel hardware.
Remote storage refers to any storage that lives on a device the user's computer is not physically connected to via a direct cable. Instead, data travels over a network. The two main on-premises forms are NAS and SAN.
A NAS is a standalone box (or set of drives in an enclosure) connected to the local network.
Every device on that network, whether a laptop, desktop, or phone, can read and write files on the NAS as though it were a local drive.
Typical use cases: shared file storage in a small office, media libraries at home, centralised backups.
NAS devices usually run a lightweight operating system and can be managed through a web interface.
A SAN is a dedicated, high-speed network purpose-built for storage traffic.
It connects servers to shared pools of block-level storage using protocols such as Fibre Channel or iSCSI.
SANs are found in data centres and large enterprises where many servers need fast, reliable access to large volumes of data.
The key distinction from NAS: a SAN presents raw block storage to the server (the server formats and manages it), whereas a NAS presents a file system that clients access directly.
Cloud storage moves data off-premises entirely, onto servers owned and operated by a third-party provider.
Well-known consumer services include Google Drive, Dropbox, and Microsoft OneDrive.
Benefits:
Accessibility – files are reachable from any device with an internet connection.
Synchronisation – changes made on one device propagate to all others.
Backup and disaster recovery – data is stored redundantly across multiple locations by the provider.
Most services offer a free tier with limited capacity and paid plans for more space.
Cloud storage depends on a reliable internet connection. Without one, access to files may be limited to whatever has been cached locally.
A smart card is a plastic card (roughly credit-card sized) with an embedded microprocessor and a small amount of memory, typically up to 64 KB.
Some smart cards also store biometric data (e.g. a fingerprint template).
Common applications:
Prepaid digital cash – stored-value cards for public transport or vending machines.
Personal identification – national ID cards, employee badges.
Access control – door locks, secure areas.
Secure transactions – chip-and-PIN payment cards, SIM cards in mobile phones.
Smart cards require a smart card reader, which may be built into a computer, a point-of-sale terminal, a door lock, or a vending machine.
The embedded microprocessor allows the card to perform on-card computation (e.g. cryptographic operations), which is what makes smart cards more secure than a simple magnetic stripe.
The chip in your bank card is a smart card. When you insert it into a payment terminal, the chip and the terminal exchange cryptographic challenges so the transaction cannot be replayed or forged. The same principle secures building access badges and government-issued ID cards.
RAID stands for Redundant Arrays of Independent Disks. The core idea is to combine multiple physical drives so they behave as one logical unit, gaining some combination of speed and fault tolerance.
RAID 0 (striping): data is split into blocks and spread across two or more disks. Because multiple disks read and write in parallel, throughput increases. The downside: zero redundancy. If any single disk fails, the entire array's data is lost.
RAID 1 (mirroring): every write goes to two disks simultaneously. If one disk fails, the mirror still holds a complete copy. Usable capacity is halved (you buy two disks but store one disk's worth of data).
Higher RAID levels (3, 5, 6, 10, etc.) combine elements of striping and mirroring, or add parity data, to balance speed, fault tolerance, and capacity efficiency. The source material focuses on RAID 0 and RAID 1 as the foundational concepts.
Large storage systems in data centres and enterprise environments typically include:
Multiple high-speed disks arranged in RAID configurations.
High-bandwidth interconnects such as Fibre Channel or iSCSI.
Scalability – the system can grow by adding more disks or shelves without replacing the whole array.
Fault tolerance and high availability are non-negotiable in these environments. A disk failure should not cause downtime or data loss.
Archival storage is for data that needs to be kept but is rarely accessed: old financial records, compliance data, completed project files, historical backups.
Large-capacity hard drives – helium-filled drives (up to 10 TB and beyond) reduce internal air resistance, allowing more platters and higher capacity in the same enclosure.
Magnetic tape – still widely used for archival. Tape cartridges and tape libraries offer extremely low cost per gigabyte and long shelf life (decades). Access is sequential, so retrieval is slow, but for data that almost never needs to be read, that trade-off is acceptable.
Optical jukeboxes – automated systems that load and read Blu-ray discs on demand. Used for media archives and regulatory retention.
Cloud-based archival – services such as AWS Glacier or Azure Archive Storage offer very low per-gigabyte pricing for data that can tolerate retrieval delays of minutes to hours.
The common thread: archival systems trade speed for cost-effectiveness and durability.
When choosing a storage solution, five factors come up repeatedly:
Speed – how fast data can be read or written. SSDs are the fastest consumer option; tape is the slowest.
Compatibility – does the storage interface (USB, SATA, Thunderbolt, wireless, network) match the device it needs to connect to?
Capacity – how much data needs to be stored now, and how much growth is expected?
Convenience and portability – a USB flash drive is easy to carry; a NAS is not.
Cost – higher capacity and faster speed usually cost more. Budget often forces a compromise.
In practice, most real-world decisions involve trade-offs. A large archival system optimises for cost and durability at the expense of speed. A laptop SSD optimises for speed and portability at the expense of raw capacity.
"NAS and SAN are the same thing." They are not. A NAS provides file-level access (clients see shared folders). A SAN provides block-level access (servers see raw disks they format themselves). NAS is simpler and cheaper; SAN is faster and more flexible for enterprise workloads.
"RAID is a backup." RAID protects against a single disk failure, but it does not protect against accidental deletion, ransomware, fire, or theft. A proper backup is a separate copy of the data, ideally stored off-site or in the cloud.
"Cloud storage means your data is on one server somewhere." In practice, cloud providers replicate data across multiple servers and often multiple data centres for redundancy. The "cloud" is a large distributed system, not a single machine.
"Smart cards store a lot of data." They typically hold up to about 64 KB, which is enough for cryptographic keys, a short biometric template, or a small balance, but nowhere near enough for files, photos, or applications.
⚠️ Know the difference between NAS (file-level, LAN-connected, simpler) and SAN (block-level, dedicated network, enterprise). This is a favourite comparison question.
⚠️ Be able to explain RAID 0 vs. RAID 1: what each does, the trade-off (speed vs. redundancy), and what happens when a disk fails under each level.
⚠️ Understand that RAID is not a substitute for backups. Examiners test this regularly.
⚠️ Cloud storage benefits (accessibility, synchronisation, backup) are commonly tested as a list. Be able to name at least three.
⚠️ The five evaluation criteria (speed, compatibility, capacity, portability, cost) may appear as a scenario question where you are given a use case and asked to recommend a storage type.
⚠️ Smart card capacity (~64 KB) and use cases (identification, access control, payment) are easy marks if you memorise them.
⚠️ Archival storage options (helium drives, magnetic tape, optical jukeboxes, cloud archival) may appear in a matching or short-answer question.
True or false: A NAS provides block-level storage to connected servers. (False – a NAS provides file-level access. A SAN provides block-level access.)
True or false: RAID 1 duplicates data across two disks. (True – that is disk mirroring.)
Fill in the blank: RAID 0 improves ________ but offers no redundancy. (speed / performance)
True or false: Smart cards typically store several gigabytes of data. (False – smart cards hold up to about 64 KB.)
Fill in the blank: Archival storage systems prioritise ________ and ________ over speed. (cost-effectiveness and durability)
Q: What is the key difference between a NAS and a SAN?
A: A NAS connects to a local network and provides file-level access (shared folders). A SAN is a dedicated high-speed network that provides block-level storage to servers, which then format and manage it themselves. NAS is simpler and suited to small/medium environments; SAN is for enterprise-scale workloads.
Q: Explain why RAID is not a backup strategy.
A: RAID protects against a single physical disk failure by using redundancy (mirroring or parity). However, it does not protect against accidental deletion, software corruption, ransomware, or physical disasters (fire, theft) that affect the entire array. A backup is a separate, independent copy of data stored elsewhere.
Q: A small business wants employees to share files from their desks. Should they use a NAS or a SAN? Why?
A: A NAS. It connects to the existing office network, provides shared folders that any device can access, and is far simpler and cheaper to set up than a SAN. A SAN would be overkill for basic file sharing.
Q: Name three benefits of cloud storage.
A: Accessibility from any internet-connected device, automatic synchronisation across devices, and built-in backup/disaster recovery through the provider's redundant infrastructure.
Q: What makes smart cards more secure than magnetic-stripe cards?
A: Smart cards contain a microprocessor that can perform on-card cryptographic operations. This means the card can authenticate itself to a reader through a challenge-response protocol, whereas a magnetic stripe simply stores static data that can be copied.
Q: A hospital needs to store 10 years of patient imaging data. Speed of retrieval is not critical, but cost and durability are. Which archival storage options would you consider?
A: Magnetic tape (very low cost per gigabyte, decades of shelf life), cloud-based archival services (e.g. AWS Glacier, with built-in redundancy), and possibly large-capacity helium-filled hard drives for on-site copies.
Q: In a RAID 0 array of four disks, what happens if one disk fails?
A: All data in the array is lost. RAID 0 stripes data across disks with no redundancy, so the failure of any single disk makes the entire array unrecoverable.
This material connects to networking modules, where NAS, SAN, and cloud architectures reappear in discussions of LAN design, protocols, and bandwidth planning. It also feeds into security and authentication topics through smart cards, cryptographic protocols, and access control. RAID and enterprise storage lead directly into systems administration and database management, where high availability and fault tolerance are everyday concerns. The evaluation criteria (speed, capacity, cost, etc.) come back in any course that covers systems design or IT procurement.
remote storage, NAS, network-attached storage, SAN, storage area network, cloud storage, Google Drive, Dropbox, OneDrive, accessibility, synchronisation, backup, disaster recovery, smart card, microprocessor, biometric, chip-and-PIN, access control, RAID, redundant arrays of independent disks, RAID 0, disk striping, RAID 1, disk mirroring, fault tolerance, high availability, Fibre Channel, iSCSI, scalability, archival storage, helium-filled drive, magnetic tape, tape library, optical jukebox, cloud archival, AWS Glacier, speed, compatibility, capacity, portability, cost, USB, SATA, wireless, storage evaluation criteria